Distributed event triggered H<inf>∞</inf> filtering with nonhomogeneous Markov switching topologies

Abstract

The distributed H ∞ state estimation problem over a filtering network with Markov switching topology is studied in this paper by employing event-triggered strategy. The strategy at each node is built on the output estimation error of its own and those received from its neighbours. Based on the communication uncertainty of practical networks, switching topology which subjects to a heterogeneous Markov chain is considered in filter design. By utilizing stochastic Markov stability theory, switching topology-dependent filters are designed such that the underlying error system is stochastically stable in mean square and the disturbance rejection attenuation level guarantees an H ∞ performance bound. An illustrative example is presented to show the applicability of the obtained results.